mistral-security-basics
Pass
Audited by Gen Agent Trust Hub on Apr 4, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements secret management best practices by discouraging hardcoded credentials and demonstrating the use of environment variables and the Google Cloud Secret Manager.
- [SAFE]: Promotes prompt injection defense through a multi-layered approach involving input length limiting, regex-based filtering of known injection patterns, and the use of XML-style boundary markers.
- [SAFE]: Integrates content moderation workflows using the official Mistral AI moderation API (mistral-moderation-latest) to check both user inputs and model outputs.
- [SAFE]: Provides output sanitization logic to mitigate XSS (script tag removal), PII leakage (SSN and email masking), and accidental system prompt disclosure.
- [SAFE]: Demonstrates secure logging practices that focus on metadata and token usage while explicitly avoiding the persistence of sensitive content or credentials.
Audit Metadata